FAKE NEWS DETECTION IN HINDI NEWS USING A PASSIVE CLASSIFIER
2021
During this modern time when people have such easy access to social media and internet, they have become very connected and sometimes it takes less than a minute for any piece of information to get viral.Consequently, fake news is spreading very easily and quickly on those platforms which lead to inappropriate actions and consequences.Sometimes this fake information is spread by people intentionally to create chaos among the people.The aim of this paper is to create a model wherein it classifies each piece of information provided to it into two categories, either non-hostile, fake, offensive or defamation.In order to accomplish the project's objective, we have suggested the algorithm term frequency -inverse document frequency.It is a statistical technique for determining how pertinent a word is in a group of documents.It is widely used for information retrieval and summarization.This method is used to calculate the term frequency of each word and further classify the words according to the occurrence in each particular type of document in the training dataset.Furthermore, the testing dataset uses this information to find the nature of the text document.This method is proposed for the natural language "Hindi" because of the unavailability of such fake news classifiers in this language.